regeneration in mrl mice- further genetic loci controlling the ear hole closure trait using mrl and...

Regeneration in MRL mice: further genetic loci controllingthe ear hole closure trait using MRL and M.m. Castaneusmice

ELLEN HEBER-KATZ, PhDa; PAN CHEN, BSa; LISE CLARK, DVM, PhDa; XIANG-MING ZHANG, BSa; SCOTTTROUTMANb; ELIZABETH P. BLANKENHORN, PhDb

The MRL mouse has been shown to display an epimorphic regenerative response after ear hole punchingleading to complete closure within 30 days and cartilage regrowth. The regenerative capacity of the MRL hasalso been seen after a severe cryoinjury to the heart leads to complete healing without scarring and func-tional myocardium. The wound healing ear hole closure response that occurs in MRL mice has been shown tobe genetically controlled. We have previously identified 11 quantitative trait loci (QTL) that govern healing inan intercross of (MRL ·C57BL/6 J) mice. However, it is desirable to use another poorly healing mouse strain toelucidate the full range of genetic factors that affect this important process. In the current study, we haveused an inbred subspecies of the mouse, M. castaneus, and have confirmed a number of loci identifiedpreviously. In addition, we report three new healing QTL. Furthermore, in this strain combination, we note astrong sexual dimorphism also observed in the MRL ·C57BL/6 cross, both in the healing trait and in the QTL thatcontrol it. (WOUND REP REG 2004;12:384–392)

The biological response to traumatic injury in higherorganisms falls into two categories: wound repair andregeneration. It is generally observed that the capacityfor tissue regeneration in mammals is limited or non-existent, especially as compared to amphibians, whereentire limbs can be regenerated after amputation.1–3

Several examples in mammals involve an ‘‘amphibian’’type of wound healing, which can be seen in there-growth and shedding of antlers4 fingertip growth5–9

and ear hole closure in rabbits.10 Ear hole closure isgenerally not seen in mammals, but there is oneinbred strain of mouse, the MRL strain, which displays fullear hole closure.11 In addition to hole closure within30 days, the MRL wound, as compared to the wound

made in a control C57BL/6 (B6) mouse strain, rapidlyreepithelializes, shows hair follicles in the wound site,and re-grows elastic cartilage. Both the fas-mutant(MRL/lpr) and the fas-wild-type MRL/MpJ strains dis-play this remarkable healing ability.

In our previous studies, we showed that ear healingwas a complex trait controlled by multiple geneticloci.12 We performed genome wide microsatellitescreens of F2 and backcross mice using MRL (healer)and B6 (nonhealer) mice as founder strains.12 Fivegenetic intervals were shown to contain quantitativetrait loci (QTL) that were significantly linked to thehealing phenotype, and several others had suggestivelinkages. Subsequently, additional wound healing QTLhave been reported, detected in experiments employingMRL and a different nonhealing parental strain, SJL/J,13

in a similar experimental design.

B6 C57BL/6

CIM Composite interval mapping

LRT Likelihood ratio test

LTBP-1 Latent TGF-b binding protein

QTL Quantitative trait loci

TGF-b Transforming growth factor-b

Original Research Articles – Regeneration Science

From the Wistar Institutea, and Department of Micro-biology and Immunologyb, Drexel UniversityCollege of Medicine, Philadelphia, Pennsylvania.

Manuscript received: August 8, 2003Accepted in final form: January 16, 2004Reprint requests: Ellen Heber-Katz, PhD, The Wistar

Institute, 3601 Spruce Street, Philadelphia, PA19104. Fax: (215) 898-3868; Email: [email protected].

Copyright # 2004 by the Wound Healing Society.ISSN: 1067-1927 $15.00 + 0

384

Knowledge of the full range of QTL that controlwound healing is dependent on, and limited by, theextent of differences between the two parental strains.For the nonhealing founders, an inbred strain that isdistantly related to MRL is ideal. The Mus musculus

group of mice consist of four subgroups14,15 includingM.m. musculus, M.m. domesticus (to which the B6,MRL, and SJL/J mouse strains belong),M.m. bactrianus,and M.m. castaneus, from which the inbred CAST/Eimouse strain is derived. It has been previously shownthat the frequency of polymorphisms between wild miceand inbred laboratory mice is much greater thanbetween different inbred strains of mice. This is espe-cially true for M. spretus; however, interbreedingdomesticus strains with spretus strains is often difficult.The frequency of polymorphisms between M. castaneus

and inbred strains is not as extensive as those seen withM. spretus, but is still sufficient to provide a more usefulset of polymorphisms for quantitative trait mapping.16

Because genetic crosses between M.m. domesticus

strains limit the genes that can be identified, a morecomprehensive understanding of genes that underlie thegenetic architecture of wound healing might be possibleby using the CAST/Ei strain as a partner for MRL. Forthis reason, we have sought to verify and extend ourknowledge of such loci by creating another segregatingcross, mating MRL mice with CAST/Ei mice to create theF1 hybrid. We expected to increase the overall numberof QTL but also to observe that certain QTL involved inwound healing would be shared between the two typesof genetic crosses. Our results suggest that this is infact a successful strategy not only for the identificationof new healing loci, but for the verification of certainhealing trait genes found in the original cross. The inter-vals containing these loci are described in this report.Further studies have revealed a highly significant differ-ence between males and females for this trait-17 andsex-specific QTL, as discussed below.

MATERIALS AND METHODSThe parental MRL/MpJ and CAST/Ei mice wereobtained from Jackson Laboratory (Bar Harbor, ME).Mice from both parental strains were bred and main-tained under standard conditions at The Wistar Insti-tute (Philadelphia, PA). F1 and F2 populations weregenerated to conduct the genetic studies. The femaleparent used for generating F1 and intercross mice wasthe MRL mouse. All animal studies were reviewed andapproved by the Wistar Institute IACUC committee.

Phenotype determinationA 2-mm through-and-through hole was made in the cen-ter of the cartilagenous part of both ears of 6-week-oldmice using a metal ear punch (Cat. #01–337B, Fisher

Scientific, Pittsburgh, PA). The holes were measured atthe time of wounding and followed for wound closureusing a grid-etched reticle (Bausch and Lomb 7·).

Genetic analysisGenomic DNA was prepared from the liver of eachanimal in the (MRL/MpJ X CASTE/Ei) · (MRL/MpJ XCASTE/Ei) F2 population. DNA from the frozen liverwas prepared using a Qiagen DNA tissue extraction kit(Qiagen Inc, Valencia, CA). We identified 138 microsa-tellite primer pairs (Research Genetics, Huntsville, AL),that were polymorphic between MRL and CAST/Ei andrelatively evenly spaced throughout the genome, andemployed them to perform a genome-wide scan of themice. Primers were radiolabeled using T4 polynucleo-tide kinase, or were used unlabeled. Amplification wasconducted with the following reagent concentrations:1· polymerase chain reaction buffer (New England Bio-labs, Beverly, MA), 0.375mM dNTPs, 0.5 U/ml of Taqpolymerase, 0.165 mM of each primer, and 160 ng/20 mlof genomic DNA. Cycling conditions include a 1 minuteat 95 ºC denaturing, 35–50 cycles of 1 minute at 94 ºC,1 minute and 30 seconds at 55 ºC, 2minutes and 10seconds at 72 ºC, and a 6-minute final extension at 72 ºC.Radiolabeled PCR products were resolved on 6 percentpolyacrylamide gels and submitted to autoradiographyas described18 and unlabeled products were run on3 percent Metaphor agarose gels (FMC, Rockland,ME) and were visualized by ethidium bromide staining.

Statistical analysisGenotype data was organized and analyzed through theuse of the Map Manager QT program.19 For quantitativetrait analysis, simple sequence repeat length poly-morphism markers were evaluated individually basedon linkage to the phenotypes, and intervals containinglikely QTL were identified by interval mapping usingthe Zmap program in Windows QTL Cartographer v1.3,model 3 (Statistical Genetics, NCSU, Raleigh, NC).Composite interval mapping (Windows QTL Cartogra-pher v1.3, model 6) was then performed on the datasetto allow for more precise definition of intervals con-taining QTL. Markers flanking the test interval areadded to the regression model to control for the pre-sence of linked QTL. Additional markers, unlinked tothe test interval but with significant effects on thetrait, are added to the model to control for the geneticbackground. The most significant markers unlinked tothe test interval are chosen using a linear regressionmodel with a forward/backward selection procedurein Windows QTL Cartographer v1.3, with a windowsize of 10 cM and the five most significant backgroundmarkers. We found this necessary because we havefound at least one example in the (MRL ·C57BL/6)F2where multiple distinct QTL were positioned on a

WOUND REPAIR AND REGENERATIONVOL. 12, NO. 3 HEBER-KATZ ETAL. 385

single chromosome (in that case, chromosome 13). Allanalyses were performed for the combined population(males and females) as well as for males and femalesseparately. Analysis for the combined population wasperformed with an additional covariate in the regres-sion model to control for differences in lesionsbetween the sexes in our population. Tests of signifi-cant linkage for a QTL are reported in the form of alikelihood ratio test (LRT) statistic. Critical values forthe declaration of significance were determined by thepermutation test routine in Windows QTL Cartogra-pher20 based on a regression model developed byChurchill and Doerge.21 The values were based ona ¼ 0.32 (one standard deviation from the mean of1000 randomly permutated LRT, as suggestive), a ¼ 0.1(strongly suggestive), a ¼ 0.05 (significant), and a ¼ 0.01(highly significant).21 LRT is equivalent to 4.6 · LOD(likelihood of linkage).

RESULTSMRL/MpJ female mice were bred to CAST/Ei malemice and the F1 hybrids tested for ear hole closure.Considerable variance in ear hole size was observed inthe (MRL ·CAST/Ei) F1. Some of this variance wasexplained by sex. Female mice were better healersthan males, with almost 50 percent being completehealers (0mm diameter after 30 days), although theaverage healing among all F1 hybrid females was notsignificantly different from that of males (0.173 � 0.193vs. 0.213 � 0.182, respectively).

The F1 mice were then intercrossed and 301 F2mice were generated and tested for their ability toheal ear wounds (Figure 1). Again, it can be seen thatthe female mice were better healers with approxi-mately 30 percent of the female mice showing completehealing. The mean residual hole size for the total(MRL·CAST/Ei)F2 population was 0.35 � 0.33mm, forthe female F2 population was 0.25 � 0.30mm, and forthe male F2 population was 0.47 � 0.32mm (differencebetween males and females, p < 1· 10�9). In addition tothe mean hole size, we analyzed the male and femaleF1 (not shown) and F2 histogram data for both skew-ness, as a measure of differences in curve symmetry,and kurtosis, as a measure of differences in data dis-tribution (Table 1). The differences in the F2 popula-tions between males and females are much morestriking than in the F1 populations.

These F2 mice were then genotyped using micro-satellite markers, at an average intermarker distance of9.4 � 6.2 cM. As QTL were identified, 20 more markerswere added to intervals of positive linkage. As can beseen in Table 2, the wound-healing trait at 30 days couldbe mapped to multiple individual loci. Linkages seen foreach marker were refined by submitting the dataset tocomposite interval mapping (CIM)22,23 (Figure 2). In this

method, potential artefactual linkages are reduced bycorrecting for linkages existing at other loci. CIMstrengthened the likelihood ratio statistic of each QTL,supporting the validity of the linkage seen at eachindividual microsatellite marker. In the total cohortof segregating progeny, individual markers or intervalsidentified by CIM showing at least suggestive linkage,using the arbitrary criterion of a ¼ 0.32, were on chro-mosomes 2, 11, 13, 14, and 17 (Figure 2). The linkage ofQTL near D13Mit16, at 10 cM on chromosome 13, wasseen in the initial (MRL · B6) F2 cross,12 and the findingof linkage in the (MRL · CAST/Ei) F2 is therefore a con-firmation of the effect of that locus, heal2, on woundhealing. The composite interval map of wound healingloci on chromosome 13 (Figure 3A) shows a broad peakwith several subpeaks, representing what are likely to beseveral independent QTL on this chromosome. This isreminiscent of the multiple QTL seen on chromosome 13in the (MRL · B6) crosses12 and likely represents a con-firmation of at least some of those QTL. One interval inparticular is located at 35 cM distal to the centromere,which showed overall significant linkage (peak LRTvalue by CIM ¼ 18.0) to healing in the MRL · CAST/EiF2: it is designated heal7.

The QTL on chromosome 11 has also been seen in areplicate study of the (MRL · B6) F2 segregating popu-lation, where it was given the designation heal10.

17 Itis located in the distal portion of the chromosome in abroad interval (60–70 cM). Another QTL was found byCIM analysis in the chromosomal region betweenD14Mit201 and D14Mit233 (from 12 to 20cM) wherethere is an interval with highly significant linkage towound healing (Figure 3C), and this is designatedheal12. The M.m. castaneus-derived allele of heal12 isresponsible for better healing and it is dominant over theMRL-derived allele at this locus.

Another highly significant linkage to the healingtrait in the total population was detected on chromo-some 17 (Figure 3D), in a location determined by CIMto be between 35 and 45 cM distal to the centromere.This QTL was highly significant in the (MRL · CAST/Ei)strain combination, with an LRS> 22, and is designatedheal13. The allele of heal13 that is associated withbetter healing is from the CAST/Ei nonhealer mouse,and behaves as an additive gene, with values for woundhealing in the heterozygotes intermediate between thetwo homozygous classes.

Analysis of the male and female populations sepa-rately (Table 2) showed that many of the QTL weresexually dimorphic. Male-associated loci included theheal2 locus on chromosome 13, and the new heal locuson chromosome 14 (Figure 3B). As in the original straincombination, heal2 is more significant to male healingthan to female healing, and shows an additive effectwith homozygous heal2

M/M mice as the best healers.A female-associated QTL was seen on chromosome 9,

WOUND REPAIR AND REGENERATION386 HEBER-KATZ ETAL. MAY–JUNE 2004

although it was of modest significance. The chromo-some 9 QTL, if confirmed in subsequent crosses, comesfrom the MRL mouse and shows an over-dominanteffect, with the heterozygotes having the most signifi-cant healing in female mice. The locus on chromosome17 showed higher significance in the male cohort,

although it is active in female mice. In general, peakLOD scores were overlapping for male and femalelinkages, although for the QTL on chromosome 11,maximum linkage to wound healing in female micemapped to a more proximal position (approximately60 cM) than that of the linkage seen in male mice

Table 1. Statistical analysis of hole diameter histograms

Total F2 Female F2 Male F2 Total F1 Female F1 Male F1

Mean 0.369 0.269 0.49 0.191 0.173 0.213Std. Dev. 0.327 0.302 0.316 0.187 0.193 0.182Std. Error 0.019 0.023 0.027 0.028 0.039 0.042Skewness* 0.707 1.36 0.187 0.912 1.023 0.831Kurtosis* �0.366 1.483 �0.852 0.587 0.611 0.759

*Skewness and kurtosis describe the histograms seen in Figure 1. Skewness is a measure of symmetry. Kurtosis is a measure of whether the data is peaked or flat relative to a

normal distribution. The female data with high kurtosis has a distinct peak near the mean and declines rapidly. The male data with low kurtosis has a flat top near the mean.

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.21.3 1.4 1.5

0

10

20

30

40

50

60

70

All

Mic

e

Hole Diameter (mm)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5

0

10

20

30

40

50

Hole Diameter (mm)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5

0

10

20

30

40

50

# of

FE

MA

LES

# of

MA

LES

Hole Diameter (mm)

(a)

(b) (c)

FIGURE 1. Ear hole closure in mapped F2 population. Three hundred and one F2 mice were ear punched and examined 30 daysafter wounding. The data presented are the number of mice (y-axis) and the given hole diameter (x-axis) at 0.1mm units measured.The three populations presented in this graph are (A) the total population; (B) the male group of the total population; and (C) thefemale group of the total population.


(approximately 70 cM). This could indicate that thechromosome 11 QTL in female mice is in fact a differ-ent QTL from the one seen on this interval in malemice, and that the resolution of heal6 may requirefurther genetic analysis of congenic mice to determineif it can be split into two QTL.

The QTL in the F2 were tested for interactions in apair-wise fashion. The chromosome 14 QTL showedvery strong epistatic and reciprocal interaction withthe chromosome 17 QTL; the additive effect of heal13was dependent on at least one copy of the MRL derivedallele of heal12 (Figure 4). This interaction indicatesthat both heal13 and heal12 could be in the same path-way and it is of interest that the pro-healing alleles forboth QTL are derived from CAST/Ei mice.

DISCUSSIONBroad-based molecular and genetic dissection of regen-eration has recently come into its own. This is due tothe ease of producing cDNA libraries, and techniquessuch as differential display and microarray analysisalong with genetic mapping studies. Such analyses havebeen carried out in animal models of regeneration suchas planaria,24 newt,25 xenopus,26,27 and the mouse.12,13,17

The potential for cross species comparison amongdiverse phyla should allow central pathways to bedefined in the future.

Our previous studies using 101 (MRL/lpr · B6) F2mice and 42 BC mice showed QTL linked to the woundhealing and regeneration phenotype on chromosomes7, 8, 12, 13, and 15. The cross in the present report,which used the Fas-normal healing strain, MRL/MpJ,and a different partner strain, CAST/Ei, confirmed thelinkage of heal2, and gave support to previously sug-gestive loci12,17 on chromosomes 11 and 13 (heal10 and

heal7). CIM of the (MRL · CAST/Ei) F2 cross docu-mented the presence of at least two significant newQTL on chromosome 14 and 17 (LRT ¼ 19.8 and 19.3,respectively). These are designated heal12 and heal13.

Therefore, MRL mice and CAST/Ei mice represent auseful, genetically different combination for this traitand illuminate new loci involved in the healing andregeneration process.

We have discovered that wound healing is con-trolled by different genes in male and female mice.Both of the QTL on chromosome 13 are more signifi-cant in male than in female mice, whereas the QTL onchromosome 9 is significant only in females, and thepooled cohort of males and females showed no linkageof this marker. This phenomenon has also beenobserved in the MRL · B6 cross17 and in other QTLanalyses of different complex traits. For example, inthe mouse model of multiple sclerosis, experimentalallergic encephalomyelitis, multiple sex-specific QTLwere found that control susceptibility and severity ofdisease in large segregating populations of (SJL/J · B10.S) F2 and backcross animals.28–30 This effecthas been seen in multiple settings, including otherautoimmune models,31 immune responses to infectiousagents,32 drug dependency,33 bone mass,34 and paintolerance.35 The broad array of different quantitativetrait models that exhibit sexual dimorphism for theirgenetic control suggests that this effect must betested in any search for quantitative genetic factorsand could be influenced by the presence of estrogenor androgens.

Interestingly, recent studies by Ashcroft andRoberts36,37 directly relate to this issue and to ourresults. We had previously demonstrated that MRLmale mice close ear holes less well than MRL femalemice and that male castration led to full healing;

Table 2. Linkage at peak markers in the (MRL/MpJ ·CAST/Ei)F2 intercross

all F2 male F2 female F2

Marker cM heal locusa

LRTb

% p-value LRT % p-value LRT % p-value

D2Mit1 1 12.9 3 0.002 8.5 5 0.014 7.6 3 0.022D9Mit182 55 NS NS 10.6 6 0.005D11Mit100/203 68/75 heal6

a 11.0 3 0.004 10.2 6 0.006 NSD13Mit16 10 heal2

a 9.7 3 0.008 7.1 4 0.028 NSD13Mit13 35 heal7

a 13.0 4 0.002 11.4 7 0.003 NSD14Mit201 12 heal12 12.0 3 0.003 14.4 9 0.001 NSD17Mit93 44 heal13 22.7* 6 1.2e-5 12.1 7 0.002 11.1 6 0.004

All markers were individually evaluated for pointwise linkage to the wound healing trait, and markers with significance at a � 0.32 were identified. Those markers with the

maximum likelihood of linkage for each interval are shown in this table. The critical values of the LRT that correspond to this value were calculated after a permutation test of the

dataset, and are shown below. The linkage evaluations in the total dataset were corrected for the difference in healing between males and females using the control routine in

MapManager QT.27

aHeal loci that were also identified in other crosses are included if the linkage p-value is < 0.05 (see text).

bLRT, likelihood ratio test statistic.cNS, not significant at a � 0.32.

Critical values Total Males Females

(a ¼ 0.32) Suggestive: 10.9 10.3 10.3

(a ¼ 0.1) Strongly suggestive: 15.4 15.2 15.9

(a ¼ 0.05) Significant: 17.2 17.4 17.6

(a ¼ 0.01) Highly significant (*): 21.2 19.2 20.8


whereas ovariectomy had no effect.17 In their woundhealing studies examining the role of smad3, a media-tor of transforming growth factor-b (TGF-b) signaling,Ashcroft et al.36,37 showed that removal of estrogen by

ovarectomy resulted in poorer wound healing in bothnormal and smad3 –/– mice; on the other hand,removal of androgens by castration resulted in betterhealing but only in normal and not in smad3 –/– mice.

CIM of chr 14

0

2

4

6

8

10

12

14

16

18

20

0 10 20 30 40 50 60 70

cM

LRT Females

Males

CIM for chr 14 QTL total cross

0

5

10

15

20

25

0 10 20 30 40 50 60 70

cM

LRT

CIM map of chromosome 13 QTL

0

5

10

15

20

0 10 20 30 40 50 60

cM

LRT

CIM for chr 17 QTL

0

5

10

15

20

25

0 10 20 30 40 50 60

cM

LRT

M

F

(a) (b) (c) (d)

FIGURE 3. Analysis of data by CIM. Using this method, potential artefactual linkages are reduced by correcting for linkages existingat other loci. CIM strengthened the likelihood ratio statistic of each QTL, supporting the validity of the linkage seen at eachindividual microsatellite marker. Plots of (A) chromosome 13 QTL; (B) chromosome 14 QTL for males and females and (C) for the totalpopulation; and for (D) chromosome 17 QTL can be seen.

LR Wound size

25

20

15

11.50

10

5

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 cM

LR Wound size25

20

15

10

11.50

5

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 cM

LR Wound size25

20

15

11.50

10

5

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 cM

(a)

(c)(b)

FIGURE 2. Depiction of the QTL identified in the (MRL·CAST/Ei) F2. Each mouse chromosome is displayed in order from its proximal todistal termini on the x-axis and the LRT on the y-axis. LRT for these composite interval map locations were generated by Windows QTLCartographer (version 1.3). The cutoff of 11.5 is a general suggestive significance indicator and seen are the LRT for the (A) total F2;(B) LRT for the male F2; and (C) LRT for the female F2.


This indicated that smad3 plays a role in androgen-mediated healing. Could this be a candidate gene?Actually, smad3 is located on chromosome 9 at 34 cMand loci mapped by both ourselves and Masinde et al.,13

(see Table 2) at this location support this molecule as apossible candidate gene. Furthermore, smad3 –/– micedisplay enhanced reepithelialization, a phenomenonseen in the MRL mouse.

We also compared results from the present cross(MRL/MpJ · CAST/Ei) F2 with previous studies usingMRL/lpr · CAST/Ei crosses used to map genes thatmodify the autoimmune phenotypes seen in the homo-zygous Fas-deficient cohort.38 None of the auto-immunity modifying loci colocalize with QTL thatcontrol the healing phenotype. This lack of corres-

pondence is also true of the MRL/lpr · B6 crosses12

and follow-up studies with this strain combination.17

Comparing our results with those of the (MRLxSJL)F2 cross reported recently,13 we find that several QTLmight be shared by both crosses: one on chromosome 4in the (MRLx CAST/Ei) F2, at approximately 50 cM withthe closest marker, D4Mit306, shows modest linkagein the point-wise marker analysis (not shown), but CIMshows the interval has suggestive level of linkage. Thisis the equivalent interval to that containing Sth4 in theMRLxSJL/J cross;13 it was weakly significant in thepublished report with MRLxB612 and in subsequentwork we have done with this strain combination.There is a broad peak on chromosome 9 with sugges-tive linkage in the female cohort of MRLxCAST/Ei F2that might align with Sth9, on chromosome 9 at 43 cM,as mentioned above. A list of all currently mappedhealing and regeneration quantitative trait loci is givenin Table 3, which shows the potential colocalizations ofthe healing trait QTL.

The intervals bearing the healing QTL discovered inthis study are too broad to allow positionally informedselection of the many candidate genes contained withinthem. However, the most significant QTL found onchromosome 17 at � 44 cM does have a very interestingcandidate gene, latent transforming growth factor betabinding protein 1 (LTBP-1).39–42 This molecule is crit-ical in TGF-b regulation and is particularly interestingconsidering a previous study showing the MRL T cellsoverproduce TGF-b.43 TGF-b is stored as a latent com-plex for future use and the LTBP is considered a mem-ber of an extracellular matrix protein superfamily thatis subject to cleavage by matrix metalloproteinasemolecules.44,45 The large latent complex consists of

Table 3. Tabulation of wound healing related loci

Origin of healer Origin of healer

Chr Name of healing QTL Peak marker cM allele in MRLxB6 allele in MRLxCAST

8 Heal1 D8Mit211 49 B613 Heal2 D13Mit116 9–13 MRL, M >F MRL, M >F13 Heal3 (Sth10) D13Mit144 50–60 MRL, M >F15 Heal4 D15Mit244 57 MRL, F >M12 Heal5 D12Mit233 13 MRL, F7 Heal6 (Sth6/7) D7Mit85 27–52 MRL13 Heal7 D13Mit245- D13Mit139 30–36 MRL, M >F MRL, M >F4 Heal8 (Sth4) D4Mit31-D4Mit148 51–71 MRL18 Heal9 D18Mit33 30–44 B6, F >M11 Heal10 D11Mit213 58–70 MRL, M >F MRL, M >F16 Heal11 D16Mit122 4–20 Heterozygote14 Heal12 D14Mit233 20 CAST/Ei, M >F17 Heal13 D17Mit93 44 CAST/Ei2 Heal15 (suggestive) D2Mit354 0–5 Heterozygote1 Sth1 D1Mit334 49.73 Sth2 D3Mit217 nd4 Sth3 D4Mit214 186 Sth5 D6Mit261 379 Sth8 D9Mit207 339 Sth9 (Heal14) D9Mit270 43 Het, F >M

Heal designations as in McBrearty et al.,12 Blankenhorn et al.,17 and this study. Sth designations as in Masinde et al.13 Where possible, the heal nomenclature has been combined

with the Sth nomenclature in intervals where there is possible overlap.

FIGURE4. Interaction between alleles at heal12 (x-axis) andheal13 (y-axis). The effect of heal13 on wound healing isdependent on the presence of at least one MRL-derived alleleof heal12.C is theM. castaneus-derived allele homozygote, H isthe heterozygote, and M is the MRL homozygote.


mature TGF-b1 homodimer, noncovalently associatedwith a disulfide-bonded complex of the N-terminalpropeptide dimer of the TGF-b1 precursor and a thirdcomponent, LTBP. Cleavage can occur via the activityof matrix metalloproteinase-2, -3, and -9, moleculesshown to be activated during MRL ear hole duringclosure.46

Another group of molecules of interest on chromo-some 14 includes angiogenin, found at 18 cM, as well asangiogenin-related protein, which may be an inactiveanalog of true angiogenin.47,48 This is especially inter-esting considering our demonstration that heart regen-eration can be seen in the MRL mouse as well asenhanced angiogenesis.49,50 Finally, bone morpho-genetic protein 4, found on chromosome 4 at 15 cM, isalso a candidate gene for regulating healing. The iden-tity of each gene underlying the QTL that control heal-ing will require the production of congenic animals thatbear the supported intervals and show a phenotypicshift in wound closure. Such studies are in progress.

ACKNOWLEDGMENTSThis work was supported by grant AI 42395 from theNational Institute of Allergy and Infectious Disease,National Institutes of Health (EHK and EPB), and theCommonwealth Universal Research EnhancementProgram, Pennsylvania Department of Health. We thankSteven Sherwood and David Moon for their assistance ingenotyping these animals. We also thank Jim Cheverudfor his useful comments.

REFERENCES1. Clark RAF. Wound repair: overview and general consideration. In:

Clark RAF, editor. The molecular and cellular biology of wound

repair. New York: Plenum Press, 1996:3–35.

2. Morgan TH. Regeneration. New York: Macmillan, 1901.

3. Stocum DL. The urodele limb regeneration blastema: determin-

ation and organization of the morphogenetic field. Differentiation

1984;27:13–28.

4. Goss RJ. Problems of antlerogenesis. Clin Orthop 1970;69:227–38.

5. Borgens RB. Mice regrow the tips of the foretoes. Science

1982;217:747–50.

6. Douglas BS. Conservative management of guillotine amputations

of the fingers of children. Aust Paediatr J 1972;8:86–90.

7. Illingworth CM. Trapped fingers and amputated finger tips in

children. J Pediatr Surg 1974;9:853–8.

8. Neufield DA. Epidermis, basement membrane, and connective-

tissue healing after amputation of mouse digits: implications for

mammalian appendage regeneration. Anat Rec 1989;223:425–32.

9. Reginelli AD, Wang YQ, Sassoon D, Muneoka K. Digit tip regen-

eration correlates with regions of Msx1 (Hox 7) expression in

fetal and newborn mice. Development 1995;121:1065–76.

10. Goss RJ, Grimes LN. Epidermal downgrowths in regenerating

rabbit ear holes. J Morph 1975;146:533–42.

11. Desquenne-Clark L, Clark R, Heber-Katz E. A new model for

mammalian wound repair and regeneration. Clin Immun Immuno-

pathol 1998;88:35–45.

12. McBrearty BA, Desquenne-Clark L, Zhang X-M, Blankenhorn EP,

Heber-Katz E. Genetic analysis of a mammalian wound healing

trait. Proc Natl Acad Sci USA 1998;95:11792–7.

13. Masinde GL, Li X, Gu W, Davidson H, Mohan S, Baylink DJ.

Identification of wound healing/regeneration quantitative trait

loci (QTL) at multiple time points that explain seventy percent

of variance in (MRL/MpJ and SJL/J) mice F2 population. Genome

Res 2001;11:2027–33.

14. Bonhomme F, Guenet J-L. The wild house mouse and its relatives.

In: Lyon MF, Searle AG, editors. Genetic variants and strains of the

laboratory mouse. Oxford: Oxford University Press, 1989:649–62.

15. Silver LM, editor. Mouse genetics. Oxford: Oxford University

Press, 1995:19.

16. Himmelbauer H, Silver LM. High resolution comparative mapping

of mouse chromosome 127. Genomics 1993;17:110–20.

17. Blankenhorn EP, Troutman S, Desquenne-Clark L, Zhang X-M,

Heber-Katz E. Sexually dimorphic genes regulate healing and

regeneration in the MRL mice. Mamm Genome 2003;14:250–60.

18. Butterfield R, Sudweeks JD, Blankenhorn EP, Korngold R,

Marini JC, Todd JA, Roper RJ, Teuscher C. New genetic loci

that control susceptibility and symptoms of experimental allergic

encephalomyelitis in inbred mice. J Immunol 1998;161:1860–7.

19. Manly KF. A Macintosh program for storage and analysis of

experimental genetic mapping data. Mamm Genome 1993;4:

303–13.

20. Basten CJ, Weir BS, Zeng ZB. QTL cartographer: a reference

manual and tutorial for QTL mapping. Raleigh: North Carolina

State University, Department of Statistics, 1997.

21. Churchill GA, Doerge RW. Empirical thresholds for quantitative

trait mapping. Genetics 1994;138:963–71.

22. Zeng ZB. Precision mapping of quanitative trait loci. Genetics

1993;136:1457–68.

23. Zeng ZB. Theoretical basis for separation of multiple linked gene

effects in mapping quantitative trait loci. Proc Nat Acad Sci USA

1994;90:10972–6.

24. Newmark PA, Sanchez Alvarado A. Not your father’s planarian: a

classic model enters the era of functional genomics. Nat Rev

Genet 2002;3:210–9.

25. Simon HG, Oppenheimer S. Advanced mRNA differential display:

isolation of a new differentially regulated myosin heavy chain-

encoding gene in amphibian limb regeneration. Gene 1996;172:

175–81.

26. Ishino T, Shirai M, Kunieda T, Sekimizu K, Natori S, Kubo T.

Identification of genes induced in regenerating Xenopus tadpole

tails by using the differential display method. Dev Dyn

2003;226:317–25.

27. KingMW, Nguyen T, Calley J, HartyMW,MuzinichMC,Mescher AL,

Chalfant C, N’ChoM,McLeaster K,McEntire J, StocumD, Smith RC,

Neff AW. Identification of genes expressed during Xenopus laevis

limb regeneration by using subtractive hybridization. Dev Dyn

2003;226:398–409.

28. Blankenhorn EP, Butterfield RJ, Rigby R, Cort L, Giambrone D,

McDermott P, McEntee K, Solowski N, Meeker ND, Zachary JF,

Doerge RW, Teuscher C. Genetic analysis of the influence

of pertussis toxin on experimental allergic encephalomyelitis

susceptibility: an environmental agent can override genetic

checkpoints. J Immunol 2000;164:3420–5.

29. Butterfield R, Blankenhorn EP, Roper RJ, Zachary JF, Doerge R,

Sudweeks JD, Rose J, Teuscher C. Genetic analysis of disease

subtypes and sexual dimorphisms in mouse experimental allergic

encephalomyelitis (EAE). relapsing-remitting and acute mono-

phasic EAE are immunogenetically distinct. J Immunol

1999;162:3096–102.

30. Butterfield RJ, Blankenhorn EP, Roper RJ, Zachary JF, Doerge RW,

Teuscher C. Identification of genetic loci controlling the character-

istics and severity of brain and spinal cord lesions in experimental

allergic encephalomyelitis. Am J Pathol 2000;157:637–45.


31. Joe B, Cannon GW, Griffiths MM, Dobbins DE, Gulko PS,

Wilder RL, Remmers EF. Evaluation of quantitative trait loci

regulating severity of mycobacterial adjuvant-induced arthritis

in monocongenic and polycongenic rats: identification of a new

regulatory locus on rat chromosome 10 and evidence of overlap

with rheumatoid arthritis susceptibility loci. Arth Rheum

2002;46:1075–85.

32. Norose K, Yano A, Zhang X-M, Blankernhorn EP, Heber-Katz E.

The mapping of genes involved in murine herpes simplex kera-

titis: identification of genes and their modifiers. J Virol 2002;76:

3502–10.

33. Buck KJ, Rademacher BS, Metten P, Crabbe JC. Mapping murine

loci for physical dependence on ethanol. Psychopharmacology

2002;160:398–407.

34. Orwoll ES, Belknap JK, Klein RF. Gender specificity in the

genetic determinants of peak bone mass. J Bone Miner Res

2001;16: 1962–71.

35. Bergeson SE, Helms ML, O’Toole LA, Jarvis MW, Hain HS,

Mogil JS, Belknap JK. Quantitative trait loci influencing morphine

antinociception in four mapping populations. Mamm Genome

2001;12:546–53.

36. Ashcroft GS, Mills SJ, Flanders KC, Lyakh LA, Anzano MA,

Gilliver SC, Roberts AB. Role of Smad3 in the hormonal modula-

tion of in vivo wound healing responses. Wound Rep Reg 2003;

11:468–73.

37. Ashcroft GS, Yang X, Glick AB, Weinstein M, Letterio JL, Mizel DE,

Anzano M, Greenwell-Wild T, Wahl SM, Deng C, Roberts AB. Mice

lacking Smad3 show accelerated wound healing and an impaired

local inflammatory response. Nat Cell Biol 1999;1:260–6.

38. Watson ML, Rao JK, Gilkeson GS, Ruiz P, Eicher EM, Pisetsky DS,

Matsuzawa A, Rochelle JM, Seldin MF. Genetic analysis of MRL-

lpr mice: relationship of the Fas apoptosis gene to disease man-

ifestations and renal disease-modifying loci. J Exp Med

1992;176:1645–56.

39. Dabovic B, Chen Y, Colarossi C, Obata H, Zambuto L, Perle MA,

Rifkin DB. Bone abnormalities in latent TGF-[beta] binding pro-

tein (Ltbp) -3-null mice indicate a role for Ltbp-3 in modulating

TGF-[beta] bioavailability. J Cell Biol 2002;156:227–32.

40. Gualandris A, Annes JP, Arese M, Noguera I, Jurukovski V,

Rifkin DB. The latent transforming growth factor-beta-binding

protein-1 promotes in vitro differentiation of embryonic stem

cells into endothelium. Mol Biol Cell 2000;11:4295–308.

41. Kulkarni AB, Thyagarajan T, Letterio JJ. Function of cytokines

within the TGF-4 superfamily as determined from transgenic and

gene knockout studies in mice. Curr Mol Med 2002;2:303–27.

42. Oklu R, Hesketh R. The latent transforming growth factor beta

binding protein (LTBP) family. Biochem J 2000;352:601–10.

43. Kench JA, Russell DM, Fadok VA, Young SK, Worthen GS, Jones-

Carson J, Henson JE, Henson PM, Nemazee D. Aberrant wound

healing and TGF-beta production in the autoimmune-prone

MRL /þ mouse. Clin Immunol 1999;92:300–10.

44. Dallas SL, Rosser JL, Mundy GR, Bonewald LF. Proteolysis

of latent transforming growth factor-beta (TGF-beta)-binding

protein-1 by osteoclasts. A cellular mechanism for release of

TGF-beta from bone matrix. J Biol Chem 2002;277:21352–60.

45. Maeda S, Dean DD, Gomez R, Schwartz Z, Boyan BD. The first

stage of transforming growth factor beta1 activation is release of

the large latent complex from the extracellular matrix of growth

plate chondrocytes by matrix vesicle stromelysin-1 (MMP-3).

Calcif Tissue Int 2002;70:54–65.

46. GourevitchD, Clark L, Chen P, Seitz A, Samulewicz S, Heber-Katz E.

Matrix metalloproteinase activity correlates with blastema forma-

tion in the regenerating MRL ear hole model. Dev Dyn 2003;226:

377–87.

47. Fu X, Roberts WG, Nobile V, Shapiro R, Kamps MP. mAngiogenin-3,

a target gene of oncoprotein E2a-Pbx1, encodes a new angiogenic

member of the angiogenin family. Growth Factors 1999;17:125–37.

48. Nobile V, Vallee BL, Shapiro R. Characterization of mouse

angiogenin-related protein. implications for functional studies

on angiogenin. Proc Natl Acad Sci USA 1996;93:4331–5.

49. Leferovich J, Bedelbaeva K, Samulewicz S, Xhang X-M, Zwas DR,

Lankford EB, Heber-Katz E. Heart regeneration in adult MRL

mice. Proc Natl Acad Sci USA 2001;98:9830–5.

50. Heber-Katz E, Leferovich J, Bedelbaeva K. Spontaneous heart

regeneration in adult MRL mice after cryo-injury. Gene Ther Reg

2002;1:399–408.


regeneration in mrl mice- further genetic loci controlling the ear hole closure trait using mrl and...

Documents